TEXTURE EVOLUTION IN A HOT ROLLED AUSTENITIC STAINLESS STEEL

The ODF analysis of the surface texture of the hot band of austenitic stainless steel reveals the 
presence of orientations of shear texture. These orientation elements are mainly distributed along two 
limited tubes of preferred orientations. The fibre of the first tube has its axis 〈110〉 ‖ RD and runs 
from {001} 〈110〉 to an orientation near {112}〈110〉 whereas the fibre of the second tube is inclined 
30° from ND towards RD (i.e. 〈110〉 30° ND fibre) and stretches from {111}〈112〉


INTRODUCTION
During hot rolling of metals, the surface undergoes a shear-strain deformation whereas the interior layers are subjected to plain-strain deformation. Depending on the conditions of rolling i.e. plain-strain or shear-strain, one obtains the rolling or shear textures in deformed metals. High surface friction, among other variables, favours the formation of shear textures near the surface of the rolled sheets (Regenet and Stiiwe, 1963, Dillamore and Roberts, 1963/64, Truszkowski, Krol and Major, 1980. In face centred cubic metals and alloys, the shear textures have been described either in terms of ideal orientations (001}(110), {112}(110), {111}(110) and {111}(112) or by the spread range between these orientations (Backofen, 1950, Backofen and Hundy, 1953, Williams, 1962. The spread range has also been described in terms of an orientation {001} (110) and a complete fibre component {111} parallel to shear plane (Hibbard, 1951, Williams, 1962, Regenet and Stiiwe, 1963. Recently Smith (1978) investigated the surface texture of a commercially produced hot band of austenitic stainless steel and observed a combination of orientations i.e. {112}(110) as major and {111}(110) as minor to characterize the surface texture. Besides experimental observations, attempts have been made to determine the shear textures theoretically from the calculations based on Taylor theory (Taylor, 1938) under plain-strain condition (Dillamore and Katoh, 1974a) and under shear-strain condition (Hansen and Mecking, 1975). In addition, the shear textures have been calculated from Taylor theory adapted to both slip and twinning (Gill Sevillano, Van Houtte and Aernoudt, 1975, Van Houtte and Aernoudt, 1976, Van Houtte, 1978. The rolling textures of face centred cubic metals and alloys possess different characters (i.e. brass type and copper type) depending on the rolling degree, rolling temperature and the stacking fault energy (SFE) and have been described in terms of two limited fibres i.e. a fibre axis (110) 11ND extending from {011}(100) to {011}(112) RD (Hirsch and Liicke, 1988). While the brass type (low SFE) is characterized by (110)II ND fibre with {011}(112) as major orientation, the copper type rolling texture forms in intermediate to high SFE metals and alloys and is described mainly by (110)60 ND fibre. The transition texture in between copper type and brass type is also perceptible. The texture transition from brass to copper type has been systematically surveyed as a function of rolling temperature in austenitic stainless steel (Goodman and Hu, 1964). Further, it has been reported by Liicke (1954) that the copper type texture is not affected much by raising the rolling temperature until the temperature is sufficiently high for recrystallization to occur. At and above such temperatures a mixture of rolling and recrystallization textures (i.e. duplex texture) forms. The well known cube texture {001} (100) forming from copper type rolling texture on recrystallization has been used along with the {021)(100) component to represent the texture at the centre of the hot band of austenitic stainless steel (Smith, 1978).
The texture characterization of the hot band of austenitic stainless steel was based on the pole figure analysis (Smith, 1978) which possesses very limited resolving power in revealing the differences which are more of quantitative than of qualitative nature. For a complete, exact and explicit description of texture, the orientation distribution function (ODF) is required, which can be obtained by the mathematical procedure of pole-figure inversion (Bunge, 1969). This paper concerns with the results of the ODF analysis of the texture near the surface and at the centre of the hot band of a commercially produced austenitic stainless steel and also discusses the process of texture evolution during hot rolling.

EXPERIMENTAL PROCEDURE
A hot band (HB) of an austenitic stainless steel of nominal composition (by wt%): 19.25% Cr, 8.40% Ni, 0.05% C, 1.24% Mn, 0.48% Si, 0.02% P, 0.02% S was commercially produced in the hot strip mill and used for the present investigation. As per supplier's information 2.95 mm thick HB strip was finished at about 1173 K and coiled at about 1023 K. The hot band was pickled in a pickling solution (by volume %) of 10% HNO3 + 2% HF + 88% distilled water at a temperature 323 K for a period of 5 minutes to remove any oxides and scales on its surface.
Optical metallography was performed on specimens mechanically polished and electrolytically etched in saturated oxalic acid solution at 10 volts for 2 minutes at sections (S 1, S 0) cut perpendicular to the transverse direction containing rolling (R) and normal (IN) direction RN. Here S denotes the thickness level at which the metallography was done and is defined as the ratio of the distance from the specimen centre to the level to the half thickness. The hardness of the HB was measured for $ 1 and S 0 levels at the sections cut parallel to the rolling plane using 5 kg load on standard Vickers hardness tester. X-ray texture measurements were carried out at the sections (i.e. S 1 and S 0 levels) parallel to the strip surface by back reflection technique using MOK radiation (Schulz, 1949). For each ground and etched specimen (20 mm x 14 mm) the texture was determined by measuring four incomplete pole figures (maximum tilt angle 75) of the plane {111), (200), (220) and {113) and plotting the ODF.
The measured intensity was subjected to background, geometrical and defocussing corrections using a random specimen of pressed and sintered austenitic stainless steel powders. The ODF was calculated from the data of four incomplete pole figures following the series expansion method of Bunge (1969) and using the pseudo-normalization technique of Kern and Bergmann (1978). The series was extended up to lmax 22 using even series only.

EXPERIMENTAL RESULTS
The commercially produced hot band of austenitic stainless steel was fully austenitic. Figure l(a) shows the optical micrograph of the HB in the RN section near the surface (S 1). The contrast in the micrograph reveals that the outer layers are not fully recrystallized and both elongated and equiaxed grains, representative of deformation and dynamic recrystallization, are present. On the other hand, the microstructure of the central layers (S 0) of the HB depicts mostly the elongated grains [Figure l(b)] and thus the interior has retained the deformed structure. Some evidence of recrystallized grains is also present. The dark bands in the above micrographs appear to be thin sheet like bands of heavily deformed austenite grains. The average grain size measured on the sections parallel to the rolling plane near the surface and at the centre of the HB was about 7 #m whereas the measured hardness values were about 235 HV5 and 269 HV5 respectively. Figure 2 shows the three dimensional orientation density distribution (i.e. ODF) of crystallites for the surface (S 1) of the HB in constant 1 sections through Euler space. This ODF exhibits maxima which are presented in Table 1 along with their Euler angles. While the table gives the exact orientations of the maxima, in the text, approximate orientations are used as indicated by the symbol to represent them by simpler indices. The maximum orientation density is exhibited by the texture component (7 7 12)(110) which is about 5 away from the ideal orientation (112} (110)  though not complete representation of the texture components. This figure depicts clearly the two limited tubes of preferred orientations in the orientation space. One of the orientation tubes has its (110) fibre axis parallel to RD and  (110)   (ll0)30ND fibre. It is evident from this figure that the density along the skeleton line increases strongly in the orientation {112}(110). The ODF for the central layer (S---0) of the HB in constant tP2 section is shown in Figure 6 and the maxima of orientations observed in this ODF along with their Euler angles are given in  (110)   (1) and Goss {011} (1) positions along (1) RD, at the centre of the hot band (HB).  (Williams, 1962). The presence of a strong {112}(110) component on the surface of Ag under the condition of high friction rolling and high reduction per pass has also been noticed (Regenet and Stiiwe, 1963 The austenitic stainless steel used in the present investigation has low SFE of the order of 38 mJm -2 (Rhodes and Thompson, 1977). The ODF analysis of the surface texture of the HB also indicated a predominant ( 112} (110) component.
It has also been reported that the maximum shear strain occurs near the surface of materials (MacGregor andCoffin, 1943, Hansen andMecking, 1975) depending on the rolling conditions and particularly under the hot rolling conditions. During hot rolling, the flow stress and work hardening exponent are lower as compared with those determined at ambient temperature and also the variable frictional forces within the length of contact of the roll gap produced by the changing speed of a metal between the entry and exit side of the mill will restrict the metal flow near the immediate surface of the metal and thus the maximum shearing strain is produced near the surface of the metal. Further, the actual hot rolling conditions in the hot strip mill are far more serious and also there is a decrease in SFE with a decreasing hot rolling temperature (Goodman and Hu, 1964). Thus, all the available information taken together, establish that all the elements of the two limited fibres (i.e. (110) [[ RD and (110)30ND) and also the orientations {111} (110) and {111} (123) observed in the surface layer of the HB of austenitic stainless steel are shear texture components. The component {129} (4 11 2) observed in the surface of the HB is associated with recrystallization texture (Musick and Liicke, 1978) and is near to {139} (341).
In the hot strip mill, during deformation, the surface layers deform at lower temperature due to the heat losses by conduction to the work rolls whereas the central layers deform at higher temperature due to the heat gain by deformation (Yanagi, 1976). For a 3 mm thick strip the difference between the temperatures of centre and surface layers is of the order of 373 K. Further during hot rolling, the SFE decreases with decreasing hot rolling temperature. The fact that the finishing temperature of the HB is about 1173K, the surface temperature during deformation is lower than that at the centre and the surface layers are partly recrystallized, would suggest that some process of simultaneous shearing and dynamic in situ recrystallization by sub grain growth to a limited extent takes place as the austenitic stainless steel is hot rolled.
On the other hand the ODF ( Figure 6) for S 0 level of the HB indicates the presence of textural elements of both the retained rolling texture (i.e. copper type) and recrystallization texture (i.e. cube and RD rotated cubes). The fibre 110)60 ND of retained rolling texture runs from {112} (111) (Virnich, K6hlhoff, Liicke et al., 1978. Eichelkraut, Hirsch and. Further, the grains of the {011}(112) component are consumed by the growth of cube grains but at a much lower rate due to unfavourable orientation relationship . The more rapid disappearance of {112} ( 111 ) component than the {123} (634) component in favour of cube orientation indicates the preferential nucleation of cube grains probably in the transition band developed within {112} (111) component. Dillamore and Katoh (1974b) based on oriented nucleation, assumed the presence of nuclei in the divergent zones. These divergent zone orientations are unstable and give rise to formation of transition bands containing the remnant of original orientation during rolling. Further, the orientations in the scattered zone between the cube and Goss positions and also many other orientations having their (100) axis close to ND come under this category. The presence of RD rotated cubes in recrystallization texture has also been interpreted as an evidence of oriented nucleation.  (112), mainly because of lower finishing temperature (1173 K) of the HB. This results in the retention of the components of copper type rolling texture and ultimately a partly recrystallized structure is produced at the centre layers of the HB.
5. CONCLUSIONS 1. The hot band (HB) of commercially produced austenitic stainless steel is partly recrystallized.
2. The extent of recrystallization is higher near the surface than at the centre.
3. At the surface of the hot band, two limited orientation tubes i.e. a fibre (110) I I RD axis running from {001} (110)